Antimicrobial activity and synergistic effect of phage-encoded antimicrobial peptides with colistin and outer membrane permeabilizing agents against Acinetobacter baumannii

Bacterial strains and growth conditions

Bacterial strains used in this study included A. baumannii ATCC19606 and twenty A. baumannii clinical isolates collected from tertiary hospitals in Thailand between November 2013 and February 2015 (Leungtongkam et al., 2018; Kongthai et al., 2021), and from April to June 2022 (this study). Bacterial strains were stored in 25% glycerol stock at −40 °C until use. Bacteria were cultivated on Leed Acinetobacter Agar (LAM) or Luria-Bertani broth (LB) incubated at 37 °C. The use of biological specimens was approved by the Naresaun University Institutional Biosafety Committee (protocol No. NUIBC MI 65-09-34).

Design of antimicrobial peptides from A. baumannii bacteriophage

Two bacteriophages, vB_AbaM_PhT2 (vPhT2) and vB_AbaAut_ChT04 (vChT04), were isolated from hospital wastewater and characterized as previously described (Kitti et al., 2014; Styles et al., 2020; Leungtongkam et al., 2023). The genome sequences of vPhT2 (MN864865) and vChT04 (OQ858591) were obtained from NCBI databases. Antimicrobial peptides consisting of 25–30 amino acid sequences were designed from target proteins of vPhT2 and vChT04. The target proteins of vPhT2 included endolysin (QHJ75684.1), tail lysozyme (QHJ75785.1), short tail fiber (QHJ75776.1), inhibitor of bacterial protein synthesis (QHJ75726.1); and endolysin (WIS40123.1) and tail tubular protein A (WIS40082.1) from vChT04 (Table 1 and Table S1). Active domains of the target protein sequences were analyzed using MOTIF search (https://www.genome.jp/tools/motif/). Antibacterial activity of peptides was predicted using AntiBP2 (Lata, Mishra & Raghava, 2010). Peptides were modified and designed using in silico methods to predict efficient cell-penetrating peptides (CPPs) as described by Gautam et al. (2013, 2015). Peptide properties were investigated using the Antimicrobial Peptide Calculator and Predictor (https://aps.unmc.edu/prediction). Antibacterial activity class and subfamily were determined using AntiBP2 (Lata, Mishra & Raghava, 2010). The three-dimensional structures of endolysin and peptides were predicted using ChimeraX (https://www.rbvi.ucsf.edu/chimerax/) and AlphaFold2 (Jumper et al., 2021). All phage peptides were synthesized by GenScript (New Jersey, USA). Lyophilized peptides were diluted in sterile ultrapure water to prepare 5 mg/ml stock solutions and stored at −20 °C for further use.

Minimum inhibitory concentration

The MICs of peptides were tested against A. baumannii ATCC19606 and clinical isolates using the broth microdilution method as per the Clinical and Laboratory Standards Institute (Clinical & Laboratory Standards Institute (CLSI), 2023). The bacteria were cultured on TSA and incubated at 37 °C overnight. Inocula were prepared by resuspending A. baumannii colonies in 0.85% NaCl solution to a concentration of 1 × 10⁵ CFU/ml using a densitometer. Two-fold serial dilutions of each peptide were prepared in Mueller-Hinton broth (MHB) in a sterile 96-well U-bottom microtiter plate. Each well was inoculated with 25 µl of bacterial solution, and the plate was incubated at 37 °C overnight. MIC was defined as the lowest peptide concentration that completely inhibited bacterial growth. Peptides with strong antimicrobial activity were further tested against MDR-AB, XDR-AB, and other bacterial species, including Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Staphylococcus aureus, and Bacillus subtilis (Table S2).

Synergism of peptides with colistin and outer membrane-permeabilizing agents

The synergistic effects of peptides with outer membrane-permeabilizing agents (benzoic acid, malic acid, lactic acid, citric acid), EDTA, and colistin were evaluated as described previously (Sitthisak et al., 2023). MICs of all agents were assessed, and 0.25× of the MIC of two agents was used in growth inhibition assays (Table S3). A. baumannii ATCC19606 was used as the host strain, and inocula were prepared by resuspending A. baumannii colonies in 0.85% NaCl solution to a concentration of 1 × 10⁵ CFU/ml. A total of 50 µl of double-strength Mueller-Hinton broth and 25 µl of bacterial inocula were added to each well of 96-well microtiter plates. Diluted peptides and test agents were added, and the plates were incubated at 37 °C overnight. After incubation, 25 µl of 0.1% triphenyl tetrazolium chloride (TTC) was added to each well and incubated for an additional hour. Absorbance was measured at 540 nm using a microplate reader. The percent inhibition was calculated using the following formula:

$$\mathrm{\%}\mathrm{i}\mathrm{n}\mathrm{h}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}={\displaystyle \frac{\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}-\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{w}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{s}\times 100}{\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}}$$

Checkerboard synergy testing was performed using peptides and test agents in 96-well plates, varying peptide concentrations across rows and tested agents across columns. A. baumannii ATCC19606 inocula (5 × 10⁵ CFU/ml) were added, and the plates were incubated overnight at 37 °C. After incubation, 25 µl of 0.1% TTC was added and incubated for an additional hour. The experiment was conducted in triplicate, and results were averaged. The Fractional Inhibitory Concentration Index (FICI) was calculated to determine synergy, with FICI ≤ 0.5 considered synergistic, FICI = 0.5–4 additive or indifferent, and FICI > 4 antagonistic (Lorian, 2005).

The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay

We aimed to determine the effects of peptides on cell viability and proliferation in the human hepatocellular carcinoma HepG2 cell line. The MTT assay was performed as described by Muangpat et al. (2022) with some modifications. The cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Sigma-Aldrich, St. Louis, MO, USA), supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin at 37 °C in 5% CO² and 95% humidity. The cells were seeded into a 96-well plate at a density of 1 × 10⁴ cells/well and allowed to adhere overnight. Then, the cells were exposed for 24 h to various concentrations of peptides. The control cells were cultured in complete DMEM medium containing 0.2% DMSO. After the 24-h incubation period, 20 μL of MTT solution (5 mg/mL in PBS) (Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) was added to each well and incubated for another 2 h at 37 °C. The culture medium was then removed, and formazan crystals formed by viable cells were dissolved in 100 μL of DMSO before measuring the absorbance at 590 nm using a microplate spectrophotometer. The effect of the peptide on cell viability was calculated as a percentage of the control, which was arbitrarily assigned a value of 100% viability. The percentage of cell viability was calculated relative to the control group. The IC₅₀ of the peptides was defined as the concentration that caused a 50% reduction in cell viability compared with the control. The MTT assay was conducted in three independent experiments, each with triplicate wells per condition.

Inhibition of biofilm formation

We determined the ability of peptides to inhibit biofilm formation in twenty biofilm-forming strains using the method described by Karyne et al. (2020) with some modifications. Briefly, biofilm-forming strains of A. baumannii were adjusted to an equivalent of the 0.5 McFarland standard in 2× Luria Bertani (LB) broth supplemented with 20% glucose. Then, 50 µL of A. baumannii and 50 µL of peptides were added. The MIC concentration of each peptide (PE04-1, PE04-1(NH₂), PE04-2) was prepared to final concentrations of 312.5, 156.25, and 156.25 µg/mL, respectively. The plates were incubated overnight at 37 °C. After 24 h of incubation, the culture medium was gently removed, and the wells were washed three times with phosphate-buffered saline. The adherent cells were fixed with absolute methanol for 10 min, stained with 0.4% crystal violet for 15 min, and washed three times with sterile distilled water, then air-dried. The plates were then filled with 250 µL of 33% acetic acid for 15 min. The absorbance at OD595 nm was determined. All experiments were performed in three independent assays, each repeated in triplicate, and the percentage of biofilm reduction was calculated based on treated wells vs. non-treated wells.

Galleria mellonella infection assay

The G. mellonella infection assay was performed as described by Peleg et al. (2009) with some modifications. Final-instar larvae weighing 180–350 mg were used. The antimicrobial activity of peptides against A. baumannii ATCC19606 and XDR-AB (AB329) was evaluated on larvae infected with 10 μL of test agents. The peptides (PE04-1, PE04-1(NH₂), PE04-2) were tested at final concentrations of 312.5, 156.25, and 156.25 µg/mL, respectively. Each peptide was prepared at 2× MIC concentration and mixed with an equal volume of 1 × 10⁸ CFU/mL bacteria. Uninfected larvae injected with 10 μL of phosphate-buffered saline (PBS) were used as control groups. Ten worms were used per group. All experimental groups were performed in triplicate. After injection, larvae were incubated at 37 °C in darkness under a humidified atmosphere with food for 10 days. Survival was recorded daily for 10 days. All experimental protocols were approved by the Naresuan University Animal Care and Use Committee (protocol No. NU-AI660602).

Statistical analyses

All experiments were independently performed in triplicate, and the data are presented as mean ± standard deviation. One-way analysis of variance (ANOVA) with Tukey’s comparison test was used to assess statistically significant differences among the experimental groups. A p value of < 0.05 was considered statistically significant. Survival data from the Galleria mellonella assays were analyzed using Kaplan-Meier survival curves, and differences between treatment groups were assessed using the log-rank test.

Characterization of the phage-encoded antimicrobial peptides

We designed ten peptides from endolysin, transcriptional inhibitor protein, tail lysozyme, and tail tubular protein A of bacteriophage vB_AbaM_PhT2 (vPhT2) (MN864865) and vB_AbaAut_ChT04 (OQ858591). The sequences and properties of all peptides are present in Table S1. We modified peptides called cell-penetrating peptides (CPPs) using amino acid replacement. PE04-2 was derived from peptide PE04-1 by amino acid substitutions (Lysine-K) at the N-terminal of peptide PE04-1 to enhance the net charge (Table S1). Additionally, an amine (NH₂) was added at the C-terminal of peptide PE04-1 to generate peptide PE04-1(NH₂). All peptides’ molecular weights, hydrophobic ratios, net charges, hydropathy values, and toxin predictions are shown in Table S1. The AMP scores of all peptides ranged from −0.185 to 1.765 (Table S1). The hydrophobic ratios of all peptides ranged from 24% to 52% (Table S1). The hydropathy values of all peptides ranged from −1.57 to 0.38 (Table S1). The sequences and predicted three-dimensional structures of PE04-1, PE04-2, and PE04-1(NH₂) peptides demonstrated in Fig. 1, showed that PE04-1 and PE04-2 formed the α-helix structure.

Minimum inhibitory concentration of phage-encoded peptides

Ten peptides derived from bacteriophage proteins were tested for antimicrobial activity. The results showed a wide range of MIC values, indicating varying levels of antimicrobial efficacy. We found three peptides, PE04-1, PE04-2, and PE04-1(NH₂), showed good activity (MIC ranging from 156.25–312.5 µg/ml), while the MICs of other peptides ranged from 1,250 to more than 5,000 µg/ml (Table 1). PE04-1, PE04-2, and PE04-1(NH₂) peptides were selected and tested against MDR-AB, XDR-AB, and other bacterial species. PE04-2 demonstrated maximum antimicrobial activity of all tested AMPs, particularly against E. coli, K. pneumoniae, and P. aeruginosa. The MICs of PE04-2 against MDR-AB and XDR-AB ranged from 78.12 to 312.5 µg/ml (Table S2). PE04-1 was less active against all Gram-negative bacteria tested in this study, while PE04-1(NH₂) was less active only in P. aeruginosa. However, all three peptides showed good antimicrobial activity against S. aureus and B. subtilis, with MIC values ranging from 19.53 to 39.06 µg/ml (Table S2). Therefore, PE04-1, PE04-2, and PE04-1(NH₂) were selected for further study.

Synergism of phage-encoded peptides with colistin and outer membrane permeabilizing agents

We performed an inhibition assay of three peptides combined with colistin, EDTA, and four weak acids (benzoic acid, malic acid, lactic acid, and citric acid) at 0.25× the MIC of two agents. The MICs of all tested agents are shown in Table S3. As shown in Fig. 2, all three peptides exhibited elevated antibacterial activity with all tested agents. The combination of peptide PE04-1 plus colistin, citric acid, EDTA, malic acid, lactic acid, and benzoic acid showed percent inhibition as 99.6%, 99.5%, 72.8%, 70.9%, 66.3%, and 40.9%, respectively (Fig. 2, Table S4). The combination of peptide PE04-1(NH₂) plus colistin, citric acid, malic acid, EDTA, lactic acid, and benzoic acid showed percent inhibition as 99.9%, 99.8%, 89.9%, 82.79%, 50.3%, 40.1% (Fig. 2, Table S4). The combination of peptide PE04-2 plus citric acid, colistin, EDTA, lactic acid, malic acid, and benzoic acid showed percent inhibition as 99.8%, 99.6%, 99.1%, 86.4%, 85.6%, and 48.7% (Fig. 2, Table S4). We performed a checkerboard assay to measure peptide synergistic effect with colistin and citric acid, which showed a high synergistic effect. The FIC indexes of PE04-1, PE04-1(NH₂), and PE04-2 plus colistin were calculated as 0.19, 0.19, and 0.14, respectively, which indicates synergism. The FIC indexes of PE04-1, PE04-1(NH₂), and PE04-2 plus citric acid were calculated as 0.37, 0.37, and 0.25, respectively, indicating synergism (Table S5).

Cytotoxicity effects of phage-encoded peptides to human hepatoma cell lines

The cytotoxicity effect of three peptides was determined using an MTT assay. All three peptides at 600 ug/ml concentrations exhibited cytotoxicity against the human liver cancer cell line (HepG2). As shown in Figs. 3A–3C, all peptides caused the death of HepG2 cell lines in a dose-dependent manner. The percentage of cell viability of the PE04-1, PE04-1(NH₂), and PE04-2 at the MIC concentration was more than 85% (Figs. 3A–3C). The concentration of the PE04-1 and PE04-1(NH₂) that caused the reduction of viable cells to 50% (IC50) was 1,440 and 525.6 ug/ml, while the IC50 of peptide PE04-2 was 424.1 ug/ml (Figs. 3A–3C). From the IC50 values, it could be suggested that concentration at MICs did not affect the cell viability of human hepatoma cell lines.

Biofilm eradication assay of phage-encoded peptides

We determined the peptides’ ability to inhibit biofilm formation using the MIC values of the peptides. Percent inhibition of all three peptides showed that PE04-1, PE04-1(NH₂), and PE04-2 showed the ability to inhibit biofilm formation against twenty A. baumannii strains with the average percent inhibition as 58.11%, 75.51%, and 91.92%, respectively as shown in Figs. 4A, 4B and Table S6. Peptides PE04-2, a CPP-modified peptide, showed strong antibiofilm effects compared to the other peptides (p < 0.05) (Fig. 4B).

G. mellonella infection assays of phage-encoded peptides

We determined peptide efficacy as a therapeutic approach in G. mellonella model. Larvae treated with PE04-1, PE04-1(NH₂), and PE04-2 at the MIC level showed survivability of XDR-AB infection of more than 80% at day 10. PE04-1(NH₂), and PE04-2 showed high survivability compared to PE04-1 (Fig. 5). Larvae injected with a dose of A. baumannii ATCC19606 showed a rapid decline in larvae number with less than 50% survival at day 5, while uninfected larvae injected with PBS or peptides, used as controls, a 100% survivability were obtained (Figs. S1–S3).